Thursday, May 17, 2012

The Problem with Psychology

Ed Yong of Not Exactly Rocket Science has just published a scathing criticism of the entire field of psychology [Replication studies: Bad copy]. The fact that this article appears in Nature should be of great concern to all psychologists. Here's the two opening paragraphs.
For many psychologists, the clearest sign that their field was in trouble came, ironically, from a study about premonition. Daryl Bem, a social psychologist at Cornell University in Ithaca, New York, showed student volunteers 48 words and then abruptly asked them to write down as many as they could remember. Next came a practice session: students were given a random subset of the test words and were asked to type them out. Bem found that some students were more likely to remember words in the test if they had later practised them. Effect preceded cause.

Bem published his findings in the Journal of Personality and Social Psychology (JPSP) along with eight other experiments1 providing evidence for what he refers to as “psi”, or psychic effects. There is, needless to say, no shortage of scientists sceptical about his claims. Three research teams independently tried to replicate the effect Bem had reported and, when they could not, they faced serious obstacles to publishing their results. The episode served as a wake-up call. “The realization that some proportion of the findings in the literature simply might not replicate was brought home by the fact that there are more and more of these counterintuitive findings in the literature,” says Eric-Jan Wagenmakers, a mathematical psychologist from the University of Amsterdam.
There's lots more where that comes from. Read the entire article.

One of the most remarkable things about Ed Yong's paper is that he doesn't even mention evolutionary psychology! I think that the absurdity of most evolutionary psychology papers is more than sufficient reason to question whether the entire field is fatally flawed [Boobies and Evolutionary Psychologists].

There's clearly something wrong. Can it be possible that an entire discipline has gone off the rails?1

1. Saying that there's a problem with a discipline is not the same as saying that there's a problem with every single psychologist. What I'm saying is that the good psychologists don't seem to have the same influence that good biochemists and good evolutionary biologist (mostly) have on their respective disciplines.


  1. How can I put this... it is not surprising to me that the field of psychiatry has plenty of bad research. A while now, it came out that the major professional organizations in the U.S. for psychiatrists and psychologists were aiding the American military in the development of pain weapons and torture methods. (And my Google-fu is failing to pull up the references, but I am sure they're there. I need caffeine.) People whose ethics will let them help design torture methods aren't going to be either smart enough or ethical enough to hold back from publishing bad science. In fact, maybe we should be glad they're muddying their own waters -- it might make them less effective at helping out with the torture and weapons.

  2. Oh, I can think of a lot better reasons that psychology is bunkum. Dozens of them. You can start with the infamous incident in which Martin Seligman, distinguished professor of psychology, etc, etc. was shocked that the lecture he gave about torturing dogs into submission given to the CIA, shortly after 9-11, was somehow misconstrued as a how-to-torture-suspects into submission lecture. Leaving me to wonder how his many decades of studying human psychology could have left him not able to imagine what any relatively bright 8th grader could have guessed would be the take-away lesson. Or you could mention his old work on torturing the gay away back when behaviorism was all the rage. And he's only one of dozens and dozens of figures in psychology that could be mentioned.

    Psychology was born making unwarranted assumptions about the nature of consciousness in order to assert that it was susceptible to scientific method, it developed some of the worst standards of research methodology and publication in something pretending to be science, authoritarian cults, etc. etc. etc. I think it's probably more responsible for more bad science and damage to the public understanding of science than any other academic field.

    When you look at the controlled research into PSI, it's, actually, far more in keeping with scientific method than huge amounts of what passes muster, for a time, as psychology. Wagenmaker's critique of Bem's work was rather effectively critiqued by Bem, Utts and Johnson, citing another critique of Wagenmaker's paper by skeptical reviwers.

    I especially liked the quote "if you pick a stupid prior, you can get a stupid posterior”.

  3. Bayesian Bouffant, FCDFriday, May 18, 2012 9:57:00 AM

    "says Eric-Jan Wagenmakers, a mathematical psychologist from the University of Amsterdam."
    A mathematical psychologist? So he studies the psychology of numbers? I was unaware that numbers had psychology.

  4. I think the whole field of ecology has the same problem.

  5. My first impression, without having read the article, is that his objections arise from the tensions between "reductive" science and "integrative" science. The above terms are, in the sense that I am using them (being unsure if someone else has used them in a closely related but fundamentally different sense), relative; that is, chemistry is "reductive" with respect to biology, psychology, and sociology and "integrative" with respect to physics.

    In my admittedly limited and biased, personal experience, the more "reductive" the science the more likely the scientists are to be reductionists and claim that the more integrative sciences are "just" their science "writ large". In instance of biology and psychology, the "problem" with psychology seems to mainly be that psychology is not biology and therefore yields "messier" results, which, while they should be reducible to biology, do not produce conclusions that are as straight-forward as biology. I am therefore left wondering if the "problem" with psychology is not so much with psychology itself but with the assumption of reductionism, which, interestingly enough, is a philosophical question and not entirely scientific.

  6. In the 90's while I was at University of Hawaii I met a man who was a psycology professor at the University of Hong Cong. I was studying physics at the time and he complained about psycologists using inductive logic instead of deductive logic. Specifically they sought papers that supported their conclusions. I was an undergraduate and not a psycology student, so do not have any right to claim expertise. But this man had some influence in my opinion of the field of psycology. It isn't a hard science. Psycologists are in a field that is so immature because we cannot define the cognative functions of the brain, psycologists kind of flail around. I respect this person he is bright and but the field he is trying to be "an expert in" is ahead of the science. Does this make sciense ;)? By the way I havn't disclosed his name to protect him.

  7. Excellent article -I've shared it with my colleagues (in a Cognitive Neuroscience lab). The drive for novel and sexy results is bizarrely strong in Psychology -to the point of journals refusing to publish "pure replications" even when they disprove important previous research! Also, the secretive nature of experimental code, data, and analysis methods/code lead to an environment that really stifles replication and, IMHO reasonable peer review.

    Also @Vicar Psychology is entirely different from Psychiatry, and faulting a scientist for what is ultimately done with their research is pretty low. Science produces knowledge, and knowledge can be used for good or evil.

    @Michael M and @Dennis. This has nothing to do with "inductive vs reductive reasoning." On the surface anyway, the method used in Psychology have been the hard-line contemporary scientific method for more than half a century now -it's the attitudes favoring novel, sexy results over replication, verification, and transparency that undermine the field.